FanPost

Pounding the Ball

Winslow Townson-USA TODAY Sports

A frequent criticism issued by some fans & media about Rajon Rondo is that he supposedly spends too much time "pounding the ball". That when he is on the floor, he spends an inordinate time dribbling the ball up slowly.

Whether this is true or not, some have also asserted that Rondo's style of play is "incompatible" with a motion offense. This is used frequently as a basketball-reason to justify proposals to trade Rondo.

In this article, I am going to assert that that (a) if Rondo is pounding the ball, it is not to any unusual or inordinate measure and that (b) it is irrelevant to whether he is capable of running a motion offense.

Now, when discussing things like this, it's helpful if we can all agree on terms.

I'm going to define the attribute of "pounding the ball" as "tendency to hold the ball during a touch for a long time before doing something with it". If that definition doesn't work for you, then I apologize. My argument here may not make sense in that case.

Next, a "motion offense" is a more technical term that in basketball jargon refers to any of several offensive schemes that rely on player motion to create (or neutralize) size advantages.

Here is a description from The Coach's Clipboard:

"A motion offense is a flexible offense that utilizes player movement, correct floor spacing, passing and cutting, and setting screens. The origin of "motion offense" is usually credited to coach Henry Iba at Oklahoma State. It was further developed and popularized by coach Bob Knight at Indiana, who utilized screening as a key part of the offense. Rather than running set plays (which can also be run in the motion offense), players move within a basic set of rules. This allows for greater flexibility than just running set plays, and will usually be effective against any kind of defense, whether man-to-man, zone or "junk" defenses. Players can move freely to open areas on the court. Once the basic concepts are learned, special patterns or plays can be designed by the coach to take advantage of his team's offensive strengths."

Now, beyond those generalities, there are several key principles that a motion offense follows. It is beyond the scope of this article to detail all of them, but three of the more prominent principles are that (a) players must not stand still, though when they move they must move with a purpose, (b) the primary goal of passing is to get the ball into the low post and (c) avoid passing to players standing still (because those are the easiest passes to pick off).

Within those principles, there are many variations of sub-strategies such as when to cut, when to set screens, whether to pass to a wing or to a high post and so on.

Reading the defense and understanding how many passes it would take to get the ball to the low post is a key calculation that all players, especially the ball-handler, have to make.

Tick Tock Look At the Clock

Let's look at the first assertion: Rondo pounds the ball too much.

It seems to me that this is something that can be measured. The game is played with a shot clock and a game clock. Video streams have frame-by-frame timestamps. If Rondo is holding on to the ball too long, then that should be an easy enough thing to just measure, right?

As it turns out, according to the nice folks at stats.nba.com, who provide data, Rondo held the ball, on average, 4.96 seconds per touch this season. Now, is that too long? Does that qualify as 'pounding the ball'?

I'm a firm believer in that, if we say something is "too long" then that must be relative to some thing that is NOT "too long". So if Rondo is holding the ball an inordinate amount, then it would seem logical that other, similar point guards would be holding it on average for much shorter time per touch.

Here are some numbers for the guards averaging the highest time-of-possession during games this season (through 81 games). These guys all average over 6 minutes of ball possession per game.

  • touches/gm == the number of touches per game the player gets.
  • touches/min == the number of touchers normalized per minute on the floor.
  • avg touch == the mean time of possession per touch in seconds.
  • %pass == the percentage of touches that end in a pass (as opposed to a shot attempt or a turnover or timeout).
  • %assOpp == the percentage of passes that result in a shot attempt
  • %touchAssOp == the percentage of touches that lead to an assist opportunity.
Player touches/gm touches/min avg touch %pass %assOp %touchAssOp
Isaiah Thomas 84.8 2.4 5.45s 70.2% 22.2% 15.6%
D. J. Augustin 77.3 2.5 5.28s 77.2% 18.4% 14.2%
Stephen Curry 78.8 2.1 5.10s 67.3% 29.2% 19.7%
Brandon Jennings 81.3 2.4 5.09s 73.4% 25.1% 18.4%
Ty Lawson 79.3 2.2 5.07s 71.5% 31.9% 22.8%
Russell Westbrook 74.7 2.4 5.06s 64.5% 26.6% 17.2%
Jordan Crawford 74.9 2.4 5.05s 75.7% 20.5% 15.5%
Mike Conley 86.3 2.6 5.01s 75.3% 18.3% 13.8%
Rajon Rondo 93.1 2.8 4.96s 78.8% 27.2% 21.4%
John Wall 95.3 2.6 4.91s 73.7% 24.6% 18.1%
Damian Lillard 84.6 2.3 4.89s 72.5% 17.9% 13.0%
Goran Dragic 77.4 2.2 4.88s 69.9% 22.6% 15.8%
Tony Parker 76.0 2.6 4.82s 73.3% 21.5% 15.8%
Jrue Holiday 81.0 2.4 4.74s 74.7% 23.6% 17.6%
Ricky Rubio 84.9 2.6 4.73s 81.5% 24.7% 20.1%
Deron Williams 77.8 2.4 4.70s 77.9% 21.0% 16.4%
Jeff Teague 77.7 2.4 4.63s 71.9% 24.7% 17.8%
Kyle Lowry 90.4 2.5 4.51s 75.9% 25.2% 19.1%
Trey Burke 79.8 2.5 4.51s 78.4% 17.4% 13.6%
Chris Paul 94.6 2.7 4.50s 75.6% 27.7% 20.9%
Kemba Walker 100.2 2.8 4.49s 77.0% 15.7% 12.1%
M. Carter-Williams 87.7 2.5 4.38s 71.4% 19.8% 14.1%
Derrick Rose 84.0 2.7 4.50s 72.3% 16.8% 12.1%
Brandon Knight 83.4 2.5 4.46s 71.1% 16.7% 11.9%

The important column is the middle one, which the table is sorted by.

Based on this, I'm just not seeing it. Rondo seems pretty average at how long he typically holds the ball. He's literally right smack in the middle of a huge cluster of guys all bunched up within 2 tenths of a second of each other.

In fact, in real, absolute terms, I'm not sure it's fair to say that any of these guys 'pound the ball' too much unless we say that they ALL do. The difference between the mean of this list and the extremes is just a half second.

Now, this is just an average time per touch. I'm sure ALL of these players have one or two plays a game where they, for whatever reason, hold possession for a much longer time. And they also have many plays where they barely hold it, moving it immediately.

What some fans tend to remember, of course, are the few notable outliers. The one or two plays where they held it for 8-10 seconds. When a fan is impatient (often for good reason) for his team to score quickly, this can seem like an eternity.

Sometimes, I imagine that a point guard has a very good reason to hold the ball an inordinate amount of time. Perhaps he's waiting for the offense to get into motion, setting up the defense. Maybe he's waiting for a pick to be set. Maybe he's winding the clock down on purpose. Maybe he's trying to draw his man to a spot. And I'm also sure there are a few times for all these players when they just make a mistake and are holding the ball longer than they should have. Whatever.

The above measures suggest that these outliers are probably occurring in similar weight for all these players.

My verdict: Rondo doesn't really 'pound the ball' any more than typical for his role.

Motion in the Ocean.

Now, let us return to the charge that Rondo's style of play is somehow 'incompatible' with a "motion offense". This charge is usually supported by the idea that a 'motion offense' involves everyone sharing the ball more equally, and that you don't want a 'ball dominant' point-guard. Rondo-critics frequently wax dreamily about the egalitarian nature of the San Antonio Spurs offense, where supposedly the ball is in state of constant motion, moving smoothly from player to player until it magically finds it's way into the basket, shot by equal chance from any of the 5 players on the floor.

Now, I have a family member who lives in San Antonio who is a big Spurs fan and so I actually kinda like the Spurs and I watch them quite a bit compared to other (non-Celtic) teams. I've managed to watch probably about a dozen of their games this season. And these wistful, fantasy-like portrayals of their style of play have often left me scratching my head.

The reality, is in every technical discussion and article I have found about the motion offense, there is absolutely nothing that asserts that the ball needs to move constantly. The required 'motion' in the motion offense is supposed to be that of the players. Certainly, there are a few specific motion offense schemes that result in fairly constant ball movement from player to player, but that is not a core principle.

The Spurs usually setup in a variation of a 4-out-1-in motion offense. This has one big in the low post and the other near the top of the three-point arc, though usually shaded to one side or the other, to balance spacing with the three wings who fill around the arc. There is no consistent strategy of loading any set pair from the 1, 2 & 3 to one side or the other. The Celtics have setup this way this year, so this should sound familiar.

In theory, any player capable of playing any of the positions can be called on to setup in any of the 5 spots or to arrive there in motion. This is what makes it an 'equal opportunity' offense. In practice, guys will not normally practice progressions that take them to positions of weakness. From these sets, the ball handler (usually Parker, though sometimes Manu) initiates motion with one of a set of initial moves with either dribble or pass. At that point all players are required to read the defense and move through a progression. The progression for each player is based on a practiced read-react strategy, but never the same in sequence. The resulting plays that are run from this are not unusual: Several variations of pick and roll/pop, motion-counter, baseline screens, etc.

Contrary to some stated conceptions on this blog by some, the Spurs offense is not designed to necessarily spread the ball or shots around equally. Tony Parker and Tim Duncan still dominate touches, each getting far more touches each game than any other teammates. And those two took by far a lot more shots than any of their teammates. Those two were the only Spurs above 25% in USG%. As for passing, Tony (31.7%) and Manu (29.2%) dominated the share of assists dished on the floor, with no other Spur posting above 17.2% AST%.

Even though the Spurs motion offense is about equal opportunity (really I would say equal _responsibility_) in positional spacing and motion, the Spurs still run a professional NBA-quality offense that properly weights usage through it's best players. Let's return to the last sentence in the quote from The Coach's Clipboard:

"Once the basic concepts are learned, special patterns or plays can be designed by the coach to take advantage of his team's offensive strengths."

It is a fundamental for optimization of any process that you want to make heavier usage of your most talented, most efficient resources and less usage of your less talented, less efficient resources.

Hence the Spurs, appropriately and to a very efficient degree, funnel their offense through their best players.

Now, nowhere in any of that is there anything that screams, "Not made for Rondo."

But wait? How can Rondo play in this style when he holds the ball so much?!!

Review the table up above. The difference in average touch possession time for Rondo and Tony Parker is 14 hundredths of a second. I suspect Tony is finding himself waiting while his teammates go in motion just like Rondo does. Yes, Rondo has been averaging more touches per game than Parker, but that is mostly a function of Tony dropping his minutes somewhat this season. When Tony was on the floor this year, he still was touching the ball 2.6 times per minute, just slightly less than Rondo's 2.8 times per minute.

In games that he played in, Parker possessed the ball for 6.1 minutes, fully 20.5% of the time he is on the floor. No other Spur was over 2.6 minutes of ball-possession time per game. Rondo possessed the ball just slightly more, at 22.9% of the time he was on the floor.

The only major difference really, is that a larger percentage of Rondo's touches (78.8%) end with a pass compared with Tony's (73.3%). The difference has been that Tony takes a couple more shots per game than Rondo. But Rondo's passes, result far more often in a shot attempt for a teammate: Some 27.2% of Rondo's passes result in 20 shot attempts per game, compared to just 12 per game for Parker (off 21.5% of his passes). In aggregate, Rondo's touches result in a shot either by himself or the person he passes to more often.

The bottom-line productivity of a point guard is point-creation: Points he scores or assists in scoring. And by that measure, Rondo and Parker were virtually indistinguishable this season, despite their very different respective teams:

Rondo scored or assisted the scoring of .377 points per touch.

Parker scored or assisted the scoring of .378 points per touch.

Now, in the spirit of the motion offense, where a pass may be to setup another pass (remember, the primary passing goal is the low post), it is important to also look at 'secondary' assist numbers. For example a play might be to dribble-drive to create a pass to the high post which then passes back in for the actual shot attempt. In this regard, Tony's passes lead to slightly higher 'secondary assists', 1.8 per game compared to 1.5 per game for Rondo. Prorated for their floor time and raw passing rates, that means a Tony Parker pass is much more likely to lead to a secondary assist.

In raw, absolute game value that is a very negligible difference. But it probably illustrates the execution quality difference between the two teams. I.E. - The Spurs now have several years of not only playing this offense, but playing it together. And, fundamentally, the Spurs are overall just a WAY more talented team. Period. Thus they are undoubtedly doing a slightly better job of making the extra pass.

I don't think I need to describe where the Celtics are right now in regards to talent and experience.

My expectation is that, with both more talented teammates and simply more and more experience together and in this offense, that the secondary assist number for Rondo would rise.

I also expect that, with more talented teammates and simply more experience together in this offense that Rondo's direct assist number will rise.

Remember those 20 shot attempts per game that Rondo's passes lead to? Those were being converted into assists this year at a 49% FG% clip.

That's good. WAY better than the Celtic's overall FG% of 43.5%, which was tied for SECOND WORSE in the NBA. If nothing else, this makes it pretty obvious that Rondo's assists are good for the shooting percentage of the team. It also suggests that Rondo was able to work effectively within this offense, despite (a) it being new for him and (b) coming off his injury and long layoff.

It is just speculation, but I can't help but wonder how many assists Rondo might have gotten if the team around him was even just 'league average' at shooting? And what if they were more experienced with playing together and in this offense?

My conclusion is that Rondo is in no way 'incompatible' with running a motion offense. In fact, I'm pretty excited to see how far he can take this offense, given more time and experience in it and with more talented teammates.

FanPosts are fan-created content and do not necessarily reflect the opinions of CelticsBlog. Be respectful and keep it clean. Thanks.